MM-131, a Bispecific Anti-Met/Epcam Mab, Inhibits HGF-Dependent and HGF-Independent Met Signaling Through Concurrent Binding to Epcam

MM-131, a bispecific anti-Met/EpCAM mAb, inhibits HGF-dependent and HGF-independent Met signaling through concurrent binding to EpCAM

Jessica B. Casalettoa, Melissa L. Geddiea, Adnan O. Abu-Yousifa, Kristina Massona, Aaron Fulghama, Antoine Boudota, Tim Maiwalda, Jeffrey D. Kearnsa, Neeraj Kohlia, Stephen Sua, Maja Razloga, Andreas Rauea,1, Ashish Kalraa, Maria Håkanssonb, Derek T. Loganb, Martin Welinb, Shrikanta Chattopadhyaya, Brian D. Harmsa, Ulrik B. Nielsena, Birgit Schoeberla, Alexey A. Lugovskoya, and Gavin MacBeatha,1

aDiscovery Division, Merrimack Pharmaceuticals, Inc., Cambridge, MA 02139; and bSARomics Biostructures AB, Medicon Village, SE-223 81 Lund, Sweden

Edited by James A. Wells, University of California, San Francisco, CA, and approved February 22, 2019 (received for review November 12, 2018) Activation of the Met receptor tyrosine kinase, either by its ligand, Given the roles of the Met signaling pathway in tumor pro- hepatocyte growth factor (HGF), or via ligand-independent mech- gression and its negative impact on clinical outcomes, Met is a anisms, such as MET amplification or receptor overexpression, has therapeutic target that remains under intense investigation. been implicated in driving tumor proliferation, metastasis, and resis- Optimal targeting of the pathway, however, requires an agent tance to therapy. Clinical development of Met-targeted antibodies that is effective in blocking both HGF-dependent and HGF- has been challenging, however, as bivalent antibodies exhibit agonis- independent signaling. Although small-molecule tyrosine kinase tic properties, whereas monovalent antibodies lack potency and the inhibitors (TKIs) designed to inhibit Met activity can, in princi- capacity to down-regulate Met. Through computational modeling, ple, achieve this, first-generation Met TKIs suffer from a lack of we found that the potency of a monovalent antibody targeting selectivity (15). Recent clinical studies with more selective Met Met could be dramatically improved by introducing a second binding TKIs have reported encouraging response rates in patients with site that recognizes an unrelated, highly expressed antigen on the MET-amplified tumors (16–18). Unfortunately, patients have

tumor cell surface. Guided by this prediction, we engineered MM-131, typically progressed relatively rapidly on Met TKIs, with an av- PHARMACOLOGY a bispecific antibody that is monovalent for both Met and epithelial erage progression-free survival (PFS) of 3.5 mo (19, 20). Notably, cell adhesion molecule (EpCAM). MM-131 is a purely antagonistic both innate and acquired resistance to TKIs has been attributed antibody that blocks ligand-dependent and ligand-independent Met to tumor microenvironment-derived production of HGF (21). As signaling by inhibiting HGF binding to Met and inducing receptor down-regulation. Together, these mechanisms lead to inhibition of pro- Significance liferation in Met-driven cancer cells, inhibition of HGF-mediated cancer cell migration, and inhibition of tumor growth in HGF-dependent Although MET is an established oncogene that drives tumori- and -independent mouse xenograft models. Consistent with its de- genesis, metastasis, and resistance to therapy, antibody ther- sign, MM-131 is more potent in EpCAM-high cells than in EpCAM- apeutics targeting Met have thus far eluded successful clinical low cells, and its potency decreases when EpCAM levels are reduced development. Met is a particularly challenging target because by RNAi. Evaluation of Met, EpCAM, and HGF levels in human tu- bivalent antibodies typically agonize Met, whereas monovalent mor samples reveals that EpCAM is expressed at high levels in a antibodies lack potency and the capacity to downregulate it. We wide range of Met-positive tumor types, suggesting a broad oppor- report the design and preclinical development of a purely an- tunity for clinical development of MM-131. tagonistic anti-Met antibody that potently blocks both ligand- dependent and ligand-independent signaling by exploiting the bispecific antibody | Met | HGF | EpCAM | cancer concept of avidity. MM-131 is a bispecific antibody that is monovalent for Met, but exhibits high avidity by concurrently binding ignaling by the Met receptor tyrosine kinase promotes pro- to the tumor-specific antigen epithelial cell adhesion molecule. Sliferation, migration, and survival, which, in turn, underlie the Preclinically, MM-131 inhibits proliferation in Met-driven cancer processes of developmental morphogenesis, wound repair, and cells, inhibits hepatocyte growth factor (HGF)-mediated organ homeostasis (1, 2). Dysregulation of Met signaling is linked cell migration, and inhibits tumor growth in HGF-dependent to cancer progression, metastasis, and resistance to therapy. Ab- and -independent mouse xenograft models. errant Met activation has been reported in many cancers and can occur via ligand-dependent and ligand-independent mechanisms. Author contributions: J.B.C., M.L.G., A.O.A.-Y., K.M., A.R., B.D.H., U.B.N., B.S., A.A.L., and The only known Met ligand, hepatocyte growth factor (HGF), can G.M. designed research; J.B.C., M.L.G., A.O.A.-Y., K.M., A.F., A.B., T.M., J.D.K., N.K., S.S., M.R., A.K., M.H., D.T.L., and M.W. performed research; J.B.C., M.L.G., A.O.A.-Y., K.M., A.F., be produced locally through autocrine and/or paracrine mecha- A.B., T.M., J.D.K., N.K., S.S., M.R., A.K., M.H., D.T.L., M.W., S.C., and B.D.H. analyzed data; nisms. For example, tumors of mesenchymal origin often pro- and J.B.C., M.L.G., T.M., J.D.K., A.R., M.H., and G.M. wrote the paper. duce their own HGF, whereas tumor-associated fibroblasts can Conflict of interest statement: Several authors are or were employees of Merrimack produce HGF to promote tumor progression in a paracrine Pharmaceuticals, Inc. at the time of contributing to the manuscript. manner (3–5). In addition to HGF-induced Met activation, This article is a PNAS Direct Submission. ligand-independent signaling can occur via MET gene amplifi- Published under the PNAS license. cation or mutation, receptor overexpression resulting from Data deposition: The atomic coordinates and structure factors have been deposited in the transcriptional up-regulation, or transactivation by other mem- Protein Data Bank, www.wwpdb.org (PDB ID codes 6HYG, for the Fc domain; 6I07, for the scFV::EpCAM complex; and 6I04, for the Fab::Met complex). Data and code for the com- brane receptors (2, 6, 7). Elevated levels of Met and/or HGF can putational model are available at http://www.data2dynamics.org/. confer resistance to therapy, including chemotherapy, radio- 1To whom correspondence may be addressed. Email: [email protected] or gavin. therapy, and targeted therapies such as EGF receptor (EGFR) [email protected]. – inhibitors (8 10). Moreover, high HGF and Met levels are as- This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. sociated with poor clinical outcomes, including increased me- 1073/pnas.1819085116/-/DCSupplemental. tastasis and decreased survival (11–14). Published online March 21, 2019.

www.pnas.org/cgi/doi/10.1073/pnas.1819085116 PNAS | April 9, 2019 | vol. 116 | no. 15 | 7533–7542 Downloaded by guest on October 2, 2021 an alternative strategy, development of therapeutic HGF/Met-blocking IgG1 CH3 domain), as well as different hinge sequences (IgG1, antibodies has also been pursued. Generation of antibodies ef- IgG4) and the addition of a flexible linker between the Fc regions. fective in blocking both ligand-dependent and ligand-independent Analysis of the resulting variants using SEC (SI Appendix, Fig. Met signaling has been difficult, however, as monovalent anti- S1A) and SDS/PAGE (SI Appendix,Fig.S1B) showed that the bodies lack potency, whereas bivalent antibodies often promote fully IgG1 variant and the IgG4/IgG1 chimera were each ∼80% receptor dimerization and activation (22, 23). To overcome these nonaggregated. To further increase antibody homogeneity and challenges, bispecific antibodies have been proposed (24, 25). Here, stability, we investigated additional changes, including using an we engineered MM-131, a bispecific antibody that is monovalent alternative knob-into-hole mutation (Y407T/T366Y), adding for Met, but has increased potency by concurrently binding the disulfide bonds, and changing the Fc linker length. The resulting tumor surface antigen, epithelial cell adhesion molecule (EpCAM). antibody variants ranged from ∼55% to ∼90% nonaggregated (SI EpCAM is a transmembrane glycoprotein, originally discovered on Appendix,Fig.S1C). Interestingly, the most improved variants from colon carcinomas, that is highly expressed in most cancers of epi- each isotype had different mutations. The best IgG1/IgG4 variant thelial origin, but is expressed at much lower levels in normal epi- (∼90% nonaggregated) contained an additional disulfide bond thelia and is spatially restricted to tight junctions (26–28). near the C terminus of CH3 and a 40-aa linker, whereas the best Here, we demonstrate that MM-131 is a purely antagonistic an- IgG1 variant (∼90% nonaggregated) contained no additional tibody that blocks both HGF-dependent and HGF-independent disulfide bonds and a 20-aa linker. Stability assays performed Met signaling in EpCAM-positive cells by inhibiting HGF binding with these two molecules, including freeze/thaw cycles, agitation, to Met and downregulating Met receptor levels. These two mecha- and elevated temperature, showed that the IgG1/IgG4 hybrid was nisms lead to the inhibition of Met-driven cancer cell proliferation more stable and slightly more nonaggregated (SI Appendix,Fig. and migration in vitro and the inhibition of HGF-dependent and S1D). Finally, we optimized the deglycosylation site using com- -independent tumor growth in vivo. Evaluation of Met, HGF, and plementary charge pairing to further enhance stability (SI Ap- EpCAM levels in human tumor samples reveals that EpCAM is pendix, Table S1). expressed at high levels in a wide range of Met-positive tumor types, To ensure that the modifications used to generate a stable, highlighting the potential for broad clinical development of MM-131. homogeneous Fc backbone did not substantially affect antibody structure, we purified and crystallized the Fc region of MM-131 Results (Fig. 1C). The structure was determined to 2.3 Å resolution by Design of MM-131. Previous studies have shown that standard, molecular replacement, with the IgG4 Fc having Protein Data bivalent antibodies targeting Met are potent antagonists in the Bank (PDB) ID code 3C2S (32) as the search model. The crys- presence of HGF, but often have the unintended and undesired tallographic data are provided in SI Appendix, Table S2. Overall, effect of activating Met in the absence of ligand. One way to cir- MM-131 Fc is more similar to a glycosylated IgG4 CH2 (rmsd = cumvent this problem is to engineer monovalent anti-Met antibodies 0.94 Å for 406 of 416 residues compared with PDB ID code (MetMAbs), such as onartuzumab (huOA-5D5) (29). Although this 4C54) (33, 34) than to an aglycosylated IgG4 CH2 (rmsd = 1.76 Å strategy reduces or eliminates antibody-induced Met activation, it for 406 of 416 residues compared with PDB ID code 4D2N) D renders the agent substantially less potent (higher EC50)thanthe (35) (Fig. 1 ). This is consistent with the intended role of the corresponding bivalent antibody and unable to induce receptor complementary charge pair introduced at position 299 of the down-regulation. One way to increase potency is to exploit the CH2 domains, which acts to restabilize the domain following concept of avidity. The functional affinity of a monovalent antibody deglycosylation. As the CH3 region of MM-131 contains knob- can, in principle, be increased simply by raising its local concentra- into-hole mutations, we compared the CH3 region of MM-131 tion at the cell surface through concurrent binding to another un- with a previously published knob-into-hole structure. As antici- related cell surface antigen. Guided by this principle, we designed pated, the CH3 region of MM-131 aligned well (rmsd = 1.59 Å MM-131 as a bispecific molecule that targets Met and the tumor for 413 of 416 residues compared with PDB ID code 4NQS) antigen, EpCAM (Fig. 1A). The engineering aspects of the bispecific (36), confirming that our design did not result in any unantici- format are discussed in the next section. EpCAM is not intended as pated structural perturbations. a therapeutic target of MM-131, but is instead used as an “anchor” To choose an anti-Met Fab, a panel of fully human anti-Met to preconcentrate MM-131 at the cell surface (EpCAM is usually antibodies was generated using a yeast-based antibody display present at much higher levels than Met). Previously, Harms et al. platform (Adimab LLC) and binned by epitope. Antibodies in (30) described a computational framework for designing and opti- both IgG and Fab formats were assayed for activation of Met in mizing bispecific antibodies to ensure efficient cross-arm binding, the absence of HGF. In the bivalent IgG format, all anti-Met given the relative expression of each target. Using this model, MM- antibodies induced phosphorylation of Met, although the degree 131 was designed to have high affinity for Met (Kd = 0.2 nM) and of agonism was epitope-dependent, with epitope bin 3 demon- SI Appendix A moderate affinity for EpCAM (Kd = 10 nM). A computational strating the least agonistic activity ( , Fig. S2 ). In simulation using these parameters, with 105 Met receptors per cell contrast, the corresponding Fabs, which are monovalent for Met, and 106 EpCAM receptors per cell, demonstrates the theoretic ef- showed no agonistic activity (SI Appendix, Fig. S2B). Using these fect of EpCAM targeting on binding avidity (Fig. 1B). data, we selected as the anti-Met portion of MM-131 a Fab from epitope bin 3 that demonstrated low aggregation properties, high Engineering and Structure of MM-131. Overall, MM-131 comprises thermal stability, and high expression yields in CHO cells. To three components: (i) a monovalent, human anti-Met Fab that identify the epitope on Met that this portion recognizes, we binds to the Sema domain of human Met; (ii) a stability- purified and crystallized a complex containing the anti-Met Fab engineered IgG1/IgG4 backbone; and (iii) a single-chain anti- of MM-131, along with the Sema domain of Met. The structure body fragment (scFv) that targets EpCAM (Fig. 1A). To generate was solved to 3.1 Å resolution, and the structure was solved by a bispecific antibody, MM-131 was engineered with an asymmet- molecular replacement using PDB ID code 4PY7 (37) as a search rical Fc region by introducing knob-into-hole mutations (31), model for the Fab and PDB ID code 4O3T (38) for Met. The rendering it heterodimeric. Size exclusion chromatography (SEC) structure of the Met Sema domain was nearly identical to pre- analysis revealed, however, that the resulting protein was only viously reported structures of Met, with an rmsd (33) from Met in 50% correctly assembled and nonaggregated (SI Appendix, Fig. PDB ID code 4O3T of 0.71 Å for 432 atoms compared. Further, S1A, sample 1). To improve antibody homogeneity, we tested the crystal structure revealed that the anti-Met Fab of MM- different antibody isotypes for the Fc region (IgG1, IgG4, and a 131 binds to propeller blades 2 and 3 of the Met Sema domain, hybrid generated by combining the IgG4 CH2 domain with the with residues from all six complementarity-determining regions

7534 | www.pnas.org/cgi/doi/10.1073/pnas.1819085116 Casaletto et al. Downloaded by guest on October 2, 2021 A Bivalent Monovalent MM-131 Anti-Met Fab K = 0.2 nM Remove pathway Add D Fig. 1. Design of a bispecific anti-Met antibody activation targeting arm cotargeting EpCAM (MM-131). (A) Schematic of the IgG4 design of MM-131. To ensure monovalency for Met, Less potent Met Increased the Fc region was engineered to include “knob-into- inhibition potency IgG1 hole” mutations, a C-terminal disulfide bond, and a Anti-EpCAM scFv linker connecting the Fc regions. (B) Simulation of K D = 10 nM Met inhibition by a monovalent, monospecific anti- Y300 body without a targeting arm (Kd = 0.2 nM) and the B 1.2 C D Bispecific increased avidity afforded by inclusion of an EpCAM 1.0 Monospecific targeting arm (Kd = 10 nM). Simulations were per- 0.8 formed assuming Met expression of 105 receptors 0.6 6 0.4 per cell and EpCAM expression of 10 receptors per Avidity 0.2 cell. (C) Ribbon diagram of the Fc region of MM-131. 0.0 (D) DE loop of the Fc domain of MM-131 (green), Fraction of free Met overlaid with the glycosylated IgG4 CH2 domain of -12 -11 -10 -9 -8 -7 -6 PDB ID code 4C54 (purple) and the aglycosylated Molecule concentration (log10 M) F296 IgG4 CH2 domain of PDB ID code 4D2N (blue). (E) E F Ribbon diagram of the anti-Met Fab of MM-131 (blue), cocrystallized with the Sema domain of Met Light chain variable region (orange). (F) Close-up view of the interface between MM-131 (blue) and the Sema domain of Met (or- 1 6 7 2 ange). (G) Ribbon diagram of the anti-EpCAM scFv of 5 3 4 MM-131 (blue), cocrystallized with the extracellular 1 domain of EpCAM (gray). The N-terminal domain of

Sema EpCAM (pyroGlu24-Leu62) is highlighted in green. (H) domain MM-131 potently inhibits HGF-mediated phosphory- of Met Heavy chain lation of Met. MM-131 was compared with mono- variable region valent anti-Met Fab and with bivalent anti-Met IgG, I PHARMACOLOGY Heavy chain both based on the anti-Met–binding domain of MM- G variable region H 1.5 Bivalent anti-Met 1.5 Bivalent anti-Met 131. Antibodies were incubated with HT-29 cells for Monovalent anti-Met Monovalent anti-Met MM-131 MM-131 2 h before 10 min of stimulation with 1 nM HGF. Ly- 1.0 1.0 HGF sates were analyzed by pMet ELISA; the dotted line pMet pMet indicates the pMet level in the absence of HGF and 0.5 0.5 antibodies (media control). (I) Similar to monovalent anti-Met Fab, MM-131 does not activate Met signal- 0.0 0.0 (Normalized to 1 nM HGF) -11 -10 -9 -8 -7 -6 -5 (Normalized to 10 nM HGF) -12 -11 -10 -9 -8 -7 -6 ing. HT-29 cells were incubated with HGF, MM-131,

Molecule concentration (log10 M) Molecule concentration (log M) Light chain 10 bivalent anti-Met, monovalent anti-Met, or media variable region for 10 min. Lysates were analyzed by pMet ELISA; the dotted lines indicate the pMet level of cells in media EpCAM alone. Plots in H and I reflect mean (n = 4) and SEM.

(CDRs) involved in binding (Fig. 1E). The antibody/antigen in- of 1.02 Å for 222 atoms compared (33). As in the previously terface has a buried surface area of 1,200 Å2 (39) and includes published structure, EpCAM contains a pyroglutamate at its N hydrogen bonds and a salt bridge (Fig. 1F). The epitope of Met terminus, rather than a glutamine. The asymmetrical unit com- that is recognized by MM-131 is in a similar region to that rec- prises an EpCAM dimer to which two monomeric scFvs are ognized by LY2875358 (the bivalent anti-Met antibody previously bound. The conformation of the dimer is in cis, which has pre- developed by Eli Lilly) (40). However, it differs substantially from viously been shown to be important for EpCAM activity (44). the epitope recognized by onartuzumab, which spans propeller Previous domain mapping has shown that most anti-EpCAM blades 4, 5, and 6 of the Sema domain (29). The epitope recog- antibodies recognize the N-terminal domain of EpCAM (45). nized by the anti-Met Fab of MM-131 overlaps with the HGF- Consistent with this, the epitope for the anti-EpCAM scFv is in β–binding site of Met (41), suggesting that MM-131 blocks binding the N-terminal domain (pyroGlu24-Leu62) (Fig. 1G). The in- of HGF-β to Met (as reported for LY2875358) (40). terface between the loops of the antibody and EpCAM is rela- Moc31 is a well-characterized anti-EpCAM antibody (42) that tively small, with a buried surface area of 878 Å2 (39). is amenable to humanization and stabilization (43). We stabi- When engineering an antibody for therapeutic use, it is im- lized moc31 onto different antibody frameworks, tested different portant to optimize the biological activity of the antibody as well orientations of the VH-VL domains, and screened for develop- as to ensure it has the appropriate biophysical properties to ability, including the percentage monomer after initial purifica- enable large-scale manufacturing and long-term storage. To test tion (SI Appendix, Fig. S3). In addition, we removed a potential the effectiveness of our bispecific antibody design, we compared site of deamidation on CDR-L1 to improve homogeneity of the MM-131 with the related monovalent anti-Met Fab (no EpCAM scFv. This change reduced the affinity of the resulting scFv, targeting) and the related bivalent anti-Met IgG (Fig. 1 H and I). shifting its equilibrium Kd from ∼3 nM to 10 nM. To determine As predicted by the model in Fig. 1B, both MM-131 and the the epitope on EpCAM recognized by this scFv, we purified and bivalent anti-Met IgG potently inhibit HGF-mediated Met crystallized the anti-EpCAM scFv complexed with the extracel- phosphorylation, whereas the monovalent anti-Met Fab is sub- lular domain of EpCAM. The structure was solved to 2.35 Å by stantially less potent (Fig. 1H). In the absence of HGF, however, molecular replacement using PDB ID codes 1SVZ for the scFv the bivalent anti-Met IgG activates Met signaling, whereas both and 4MZV for EpCAM. Overall, the structure of EpCAM in the the monovalent Fab and MM-131 show no discernible agonistic scFv::EpCAM complex is very similar to a previously published activity (Fig. 1I). To assess the stability of MM-131, we developed a structure of EpCAM (PDB ID code 4MZV) (44), with an rmsd purification scheme with resins suitable for large-scale manufacturing,

Casaletto et al. PNAS | April 9, 2019 | vol. 116 | no. 15 | 7535 Downloaded by guest on October 2, 2021 protein A and hydroxyapatite, resulting in material that was Dual Mechanism of Action of MM-131. MM-131 contains an anti- 97.8% monomeric (SI Appendix, Fig. S4A). After 14 mo at 4 °C, Met Fab that binds to propeller blades 2 and 3 of the Met Sema the antibody was 96.2% monomeric, demonstrating its potential domain, overlapping the HGF-β–binding site. This is similar to the for long-term storage (SI Appendix, Fig. S4B). epitope reported for the bivalent anti-Met antibody LY2875358, but dissimilar from that of huOA-5D5, which binds to propeller EpCAM Targeting Enhances Blockade of Downstream Signaling. To blades 4, 5, and 6 and is reported to block HGF-α binding (29, 40). assess the effects of MM-131 on downstream signaling, as well as In contrast, the bivalent anti-Met antibody ABT-700 binds an to determine how different levels of Met and EpCAM affect epitope outside the Sema domain and is reported to block Met MM-131 activity, we built a mechanistic model of Met signaling dimerization (46). To confirm that MM-131 is HGF-competitive, comprising ordinary differential equations. Each equation de- biolayer interferometry was used to measure the HGF-blocking scribes either a protein-binding or catalytic event in the network, properties of MM-131, LY2875358, huOA-5D5, and ABT-700. and the model includes intracellular signaling pathways and receptor Antibodies were loaded onto anti-human IgG Fc capture bio- trafficking. Model simulations suggested that MM-131 would pro- sensors, followed by sequential incubation with 200 nM human vide enhanced inhibition of pAkt relative to the monovalent anti- Met, 200 nM human HGF, and PBS. As expected, MM-131, body, huOA-5D5, across a wide range of Met and EpCAM surface huOA-5D5, and LY2875358 all blocked HGF binding to Met, receptor levels (Fig. 2A). Importantly, most cell lines derived from whereas ABT-700 did not (Fig. 3A). By blocking HGF binding to epithelial cancers fall into an area of this space that strongly favors Met, MM-131 prevents HGF-mediated Met phosphorylation and MM-131 over huOA-5D5 (Fig. 2A and SI Appendix,TableS3). activation of Gab1, Akt, MAPK (Erk1/2), and S6 in a dose- Using this model, three cell lines were selected that lie in regions dependent manner (Fig. 3B). where MM-131 is predicted to show either enhanced (NCI- In addition to HGF-induced Met activation, Met can be acti- H747 and NCI-H441) or nearly equivalent (A549) activity relative vated by HGF-independent mechanisms, including gene amplifi- to huOA-5D5 (Fig. 2A). Cells were then incubated with either MM- cation, mutation, and transcriptional up-regulation. To block both 131 or huOA-5D5, stimulated with 1 nM HGF for 10 min, and HGF-dependent and HGF-independent Met signaling, MM-131 analyzed for pAkt (S473) by ELISA (Fig. 2B and SI Appendix,Fig. was designed to down-regulate the receptor. Antibody-mediated S5A). MM-131 was more potent than huOA-5D5 in cells with a receptor down-regulation, which occurs via receptor internaliza- high EpCAM/Met ratio and equipotent in cells with a low EpCAM/ tion and degradation, typically requires bivalent binding; thus, Met ratio. Further validation was carried out in isogenic cell lines in huOA-5D5 does not induce this effect. In contrast, MM-131 which EpCAM was stably knocked down by RNAi. As predicted, induces down-regulation of Met in both MET-amplified (SI Ap- lowering the EpCAM level (and hence the EpCAM/Met ratio) in pendix, Fig. S6A) and nonamplified (Fig. 3C) EpCAM-positive NCI-H747 and NCI-H441 cells rendered MM-131 less potent at cells in the absence of HGF. The magnitude of this effect de- inhibiting HGF-mediated Met pathway activation (Fig. 2C and SI pends on the bivalent binding of MM-131 to EpCAM; less ef- Appendix,Fig.S5B and Table S4). fective down-regulation of Met was observed using a version of

A Comparision of potency between MM-131 pand huOA-5D5 103 NCI-H441 NCI-H747 Fold-change in IC90 inhibion of Akt phosphorylation (Values >1 indicate that MM-131 is favorable)

100-fold 107

60-fold

106 40-fold

A549 20-fold 105

10-fold EpCAM level (molecules/cell) 103 6-fold

4-fold 103 104 105 106 Met level (molecules/cell) B A549 NCI-H747 C NCI-H747 1.2 MM-131 1.2 MM-131 1.2 shControl + MM-131 1.0 huOA-5D5 1.0 huOA-5D5 1.0 shEpCAM + MM-131 0.8 0.8 0.8 shControl + huOA-5D5

0.6 0.6 0.6

pAkt 0.4 pAkt 0.4 pAkt 0.4 0.2 0.2 0.2 0.0 0.0 0.0 -11 -10 -9 -8 -7 -6 -11 -10 -9 -8 -7 -6 -11 -10 -9 -8 -7 -6 (Normalized to 1 nM HGF) (Normalized to 1 nM HGF) (Normalized to 1 nM HGF)

Molecule concentration (log10 M) Molecule concentration (log10 M) Molecule concentration (log10 M)

Fig. 2. EpCAM targeting increases the potency of MM-131. (A) Computational simulation comparing the potency of MM-131 with that of huOA-5D5. For

each level of Met expression (x axis) and EpCAM expression (y axis), the IC90 for inhibition of Akt phosphorylation was calculated at 10 min following HGF stimulation. The values in the heat map indicate the fold change between MM-131 and huOA-5D5 (values >1 indicate that MM-131 is favorable). Dots represent a panel of cell lines with different cell surface levels of Met and EpCAM measured by quantitative fluorescence-activated cell sorting. (B) Testing model predictions in cell lines with low and high ratios of EpCAM/Met (0.33 for A549 and 28.9 for NCI-H747). (C) Testing model predictions in an isogenic cell line in which EpCAM was stably knocked down by RNAi. The EpCAM/Met ratio in NCI-H747 shEpCAM is 1.74. Antibodies were incubated with the indicated cell lines for 2 h before 10 min of stimulation with 1 nM HGF. Lysates were analyzed by pAkt ELISA; the dotted lines indicate the pAkt level in the absence of HGF and antibodies (media control). Plots in B and C reflect mean (n = 4) and SEM.

7536 | www.pnas.org/cgi/doi/10.1073/pnas.1819085116 Casaletto et al. Downloaded by guest on October 2, 2021 ABMet HGF Buffer MM-131 C 2.0 -HGF+HGF EpCAMmutMetmut pMet (Y1234/1235)

1.5 Met PBS PBS MM-131MM-131MM-131huOA-5D5LY2875358 1.0 pGab1 (Y659) Met Gab1 0.5 EpCAM Binding (nm) pAkt (S473) Histone H3 0.0 Akt 0 100 200 300 400 0 h 24 h Time (sec) pErk1/2 (T202/T204) Erk1/2 120 100 PBS MM-131 pS6 (S235/236) huOA-5D5 80 MM-131 S6 EpCAMmut 60 MM-131 LY2875358 Metmut Actin MM-131 ABT-700 40 huOA-5D5 20 LY2875358

Percent remaining 0 Met EpCAM

Fig. 3. MM-131 inhibits HGF-dependent and HGF-independent Met signaling by antagonizing HGF binding to Met and inducing down-regulation of Met. (A)MM- 131 blocks HGF binding to Met. Biolayer interferometry (ForteBio) was used to measure the HGF-blocking properties of anti-Met antibodies MM-131, huOA-5D5, LY2875358, and ABT-700. Antibodies were loaded onto anti-human IgG Fc capture biosensors and incubated with 200 nM human Met. After 2 min, 200 nM human HGF wasadded,followedbyadditionofbufferat4min.(B) MM-131 inhibits HGF-induced Met signaling in a dose-dependent manner. HT-29 cells were incubated without HGF, with 1 nM HGF, or with 1 nM HGF + 0.1–1,000 nM MM-131 for 10 min and analyzed by immunoblotting with the indicated antibodies. (C) MM-131 down- regulates total Met levels. NCI-H2170 cells were incubated with 100 nM indicated antibodies or PBS for 24 h and analyzed by immunoblotting. Histone H3 served as a loading control. Bands were quantified using the LI-COR Odyssey imaging system. MM-131EpCAMmut is a version of MM-131 with a mutation in the EpCAM scFv, preventing EpCAM binding. MM-131Metmut is a version of MM-131 that cannot bind Met. Data are the mean of two independent samples; error bars indicate SD.

MM-131 that contains a mutation in CDR-H1 of the anti-EpCAM MM-131 Inhibits HGF-Dependent and HGF-Independent Tumor Growth scFv that impairs EpCAM binding (MM-131EpCAMmut)(Fig.3C in Vivo. To determine whether the antiproliferative activity of and SI Appendix,Fig.S6A). Together, the dual mechanism of action MM-131 observed in vitro translates to similar activity in vivo, we of MM-131 suggests that MM-131 will block both HGF-dependent evaluated MM-131 in several xenograft models. As murine HGF and HGF-independent Met signaling in EpCAM-positive tumors. does not activate human Met (48), we examined the activity of

MM-131 in tumors with HGF-dependent Met activation by sta- PHARMACOLOGY MM-131 Inhibits Cancer Cell Proliferation and Migration. To de- bly expressing human HGF in NCI-H358 and HCC827 lung termine whether the inhibition of Met signaling by MM-131 has cancer cell lines before implantation. Treatment with 12 mg/kg phenotypic consequences, we examined the ability of MM-131 to of MM-131 prevented growth of NCI-H358–HGF and HCC827- inhibit cancer cell proliferation in cell lines with HGF-dependent HGF tumors (Fig. 5A and SI Appendix, Fig. S8A, respectively). or HGF-independent Met activation. Cell viability was assayed in These results are consistent with the mechanism of action 3D spheroids following 96 h of treatment with MM-131, MM- of MM-131 (Fig. 3 A and C), the inhibition of HGF-induced 131EpCAMmut, huOA-5D5, or LY2875358. As suggested by the Met signaling by MM-131 (Fig. 3B), and the inhibition of blockade of Met signaling, MM-131 inhibits HGF-dependent HGF-mediated proliferation by MM-131 in vitro (Fig. 4A and SI Appendix A proliferation of Met-positive cells (Fig. 4A and SI Appendix, , Fig. S7 ). Notably, treatment with 10 mg/kg of huOA- Fig. S7A) and HGF-independent proliferation of MET-amplified 5D5 or 15 mg/kg of LY2875358 (molar equivalents of 12 mg/kg cells (Fig. 4B and SI Appendix, Fig. S7B) in a dose-dependent of MM-131) was less effective at inhibiting tumor growth than manner. Consistent with the decrease in MM-131 potency ob- treatment with MM-131, although the difference between MM- served in EpCAM knockdown cell lines [short hairpin EpCAM 131 and LY2875358 was generally modest. C SI Appendix B The antiproliferative effect of MM-131 on MET-amplified cells (shEpCAM); Fig. 2 and , Fig. S5 ] and the in- B SI Appendix B complete down-regulation of Met by MM-131EpCAMmut (Fig. 3C (Fig. 4 and ,Fig.S7 ) also translated to tumor growth inhibition in MET-amplified SNU-5 and MKN-45 gastric cancer and SI Appendix, Fig. S6A), MM-131 is less effective at inhibiting and NCI-H1993 lung cancer xenograft tumors. MM-131 inhibited proliferation in shEpCAM cells, and MM-131EpCAMmut does not tumor growth more effectively than huOA-5D5 and LY2875358 in effectively inhibit proliferation in control cells. each model (Fig. 5B and SI Appendix,Fig.S8B). Moreover, in the Given the role of HGF/Met signaling in tumor invasion and SNU-5 model, a complete response was observed in 100% (seven metastasis (47), we also examined the ability of MM-131 to inhibit of seven) of MM-131–treated animals, 71% (five of seven) of HGF-mediated cancer cell migration in a wound-healing assay. Cells LY2875358-treated animals, and 0% (zero of seven) of huOA- were grown to confluence; scratched with a 96-pin wounding device; 5D5– or vehicle-treated animals (SI Appendix,Fig.S8C). Evalua- and treated with MM-131, huOA-5D5, or LY2875358 in the pres- tion of total Met and phospho-Met levels in SNU-5 xenografts ence or absence of HGF. Images of the wound were captured at 2-h revealed significantly decreased Met and pMet levels in MM-131– intervals for 24 h, and migration rates were calculated from these and LY2875358-treated animals, whereas Met and pMet levels data. Whereas treatment with huOA-5D5 or LY2875358 allows were unchanged following treatment with huOA-5D5 and vehicle migration and wound closure by 24 h, MM-131 inhibits HGF-mediated (Fig. 5 C and D). These results are consistent with the mech- C E SI Ap- migration in a dose-dependent manner (Fig. 4 and and anism of action of MM-131 and LY2875358 (blocking HGF pendix C EpCAMmut ,Fig.S7 ). MM-131 slows migration and prevents binding and downregulating Met) and the inability of huOA-5D5 to wound closure but is less effective than MM-131. Notably, in the down-regulate Met, and thus inhibit growth of MET-amplified absence of HGF, huOA-5D5 and LY2875358 promote migration tumors. Consistent with tumor regression in MM-131– and and wound closure, whereas MM-131 does not (Fig. 4D and SI Ap- LY2875358-treated animals, the percentage of Ki67-positive pendix,Fig.S7D). Together, these data demonstrate that MM-131 tumor cells was lower in these tumors (Fig. 5D). inhibits cancer cell proliferation in cell lines with HGF-dependent To determine whether EpCAM targeting of MM-131 is critical or HGF-independent Met activation and blocks HGF-mediated for the inhibition of tumor growth in vivo, we stably knocked cancer cell migration. In addition, EpCAM targeting of MM-131 down EpCAM in NCI-H441 lung and MKN-45 gastric cancer is critical for maximal inhibition of HGF/Met-driven proliferation cells by RNAi (shEpCAM) and evaluated MM-131 in NCI-H441 and migration. and MKN-45 shEpCAM and shControl xenografts. Treatment

Casaletto et al. PNAS | April 9, 2019 | vol. 116 | no. 15 | 7537 Downloaded by guest on October 2, 2021 AB NCI-H747 MKN-45 1.2 1.2 shControl + MM-131 1.0 1.0 shEpCAM + MM-131 EpCAMmut 0.8 0.8 shControl + MM-131 shControl + huOA-5D5 0.6 0.6 shControl + LY2875358 Relative viability

Relative viability 0.4 0.4 (Normalized to control) 0.2 (Normalized to 1 nM HGF) -11 -10 -9 -8 -7 -6 -5 -10 -9 -8 -7 -6

Molecule concentration (log10 M) Molecule concentration (log10 M) C D 1.5 NCI-H441 3.0 NCI-H441

1.0 2.0

0.5 1.0

0.0 Normalized migration rate 0.0 Normalized migration rate -9 -8 -7 -6 -9 -8 -7 -6

Molecule concentration (log10 M) Molecule concentration (log10 M) E T=0 h T=24 h Vehicle Vehicle huOA-5D5 LY2875358 MM-131EpCAMmut MM-131

Fig. 4. MM-131 inhibits cancer cell proliferation and migration. MM-131 blocks HGF-dependent cell proliferation (A) and proliferation of MET-amplified cells (B) in a dose-dependent manner. NCI-H747, NCI-H441, and MKN-45 cell lines were stably transduced with lentiviral particles expressing untargeted shRNA (shControl) or shRNA targeting EpCAM (shEpCAM). Following formation of 3D spheroids, cells were treated with the indicated antibodies for 96 h. Viability was assessed by CellTiter-Glo assay and normalized to vehicle-treated cells in the presence (A) or absence (B) of 1 nM HGF. (C) MM-131 blocks HGF-induced cell migration in a dose-dependent manner. Following production of a uniform scratch wound, images were taken every 2 h for 24 h. Migration rates were calculated from the relative wound densities (density of cells inside the wound area/density outside the wound area) of each image and normalized to vehicle-treated cells in the presence of 1 nM HGF. (D) Unlike huOA-5D5 and LY2875358, MM-131 does not promote migration in the absence of HGF. A scratch wound migration assay was performed as in C, except in the absence of HGF. Migration rates were normalized to vehicle-treated cells in the absence of HGF. (E) Representative images of NCI-H441 cells from an HGF-induced migration assay. Images are from 1 μM treatment of indicated molecules. The wound area remaining is shown in gray, and the confluence mask is shown in orange. All plots reflect mean (n = 3) and SEM.

with MM-131 resulted in a marked reduction of tumor growth in migration in models driven by Met signaling, and that MM-131 is shControl xenografts but was not effective in shEpCAM cells (SI most effective in cells that also express EpCAM. To determine Appendix, Fig. S8D), demonstrating that EpCAM targeting in- the frequency of Met, EpCAM, and HGF coexpression in tumors, creases the activity of MM-131 in vivo. Notably, the activity of serial sections of primary tumor samples from 93 patients with co- MM-131 in NCI-H441 and MKN-45 shEpCAM xenografts mir- lorectal cancer (CRC), 99 patients with gastric cancer, and 97 patients rored that of huOA-5D5, highlighting the lack of efficacy of with non-small cell lung cancer (NSCLC) were stained for each monovalent anti-Met therapeutics in this setting and the neces- biomarker. Expression of Met was observed in 77%, 68%, and 47% sity of the EpCAM-targeting arm of MM-131. of CRC, gastric cancer, and NSCLC samples, respectively (Fig. 6). EpCAM was detected in nearly all samples (93%, 97%, and 92% in MM-131 Has No Observed Toxicity in Cynomolgus Monkeys. The Met CRC, gastric cancer, and NSCLC, respectively), and coexpression of and EpCAM epitopes recognized by MM-131 are well conserved Met and EpCAM was observed in 96% of samples that scored pos- between humans and cynomolgus monkeys (Macaca fascicularis). itive for Met across all indications. Further work will be necessary to Accordingly, MM-131 binds with similar affinities to the human gain a more quantitative understanding of EpCAM and Met levels in and cynomolgus monkey versions of both antigens (SI Appendix, both tumor samples and normal tissue. Expression of HGF in Met- Table S5). To determine the maximum tolerated dose of MM- positive samples was observed in 44% (81 of 185) of samples. Overall, 131, monkeys were dosed weekly with 10 mg/kg, 30 mg/kg, or 43% (79 of 185) of Met-positive tumors coexpressed EpCAM and 60 mg/kg of MM-131 via i.v. infusion for 4 wk. No MM-131–related HGF, suggesting a broad clinical opportunity for MM-131. changes in clinical pathology parameters, histopathological changes, or gross pathological changes were observed at any of Discussion the dose levels. Thus, the no-observed-adverse-effect level of this Aberrant Met signaling plays a key role in cancer initiation, study was the highest dose tested, or 60 mg/kg. Repeat dose progression, metastasis, and resistance to therapy. In some pharmacokinetic studies in monkeys revealed a serum half-life that cancers, Met behaves as a driver; tumors with MET gene am- ranged from 12.5 to 72.4 h. Assuming MM-131 displays a longer plification or mutation are often “addicted” to Met signaling for half-life in humans than monkeys, dosing once every three half-lives their proliferation and survival (6). MET amplification has been could potentially support biweekly dosing in a clinical setting. reported in ∼2% of epithelial cancers, including gastroesophageal cancers, CRC, and NSCLC, and is one of the mechanisms re- Met, EpCAM, and HGF Are Coexpressed in Tumors. Preclinical data sponsible for acquired resistance to EGFR therapy in NSCLC and demonstrate that MM-131 inhibits cancer cell proliferation and CRC (12, 13, 49–52). More often, however, Met is overexpressed

7538 | www.pnas.org/cgi/doi/10.1073/pnas.1819085116 Casaletto et al. Downloaded by guest on October 2, 2021 A NCI-H358-HGF B SNU-5 1600 ) 500 ) 3 3 400 1200 Vehicle huOA-5D5 300 ** 800 LY2875358 *** 200 *** MM-131 * **** 400 100 Tumor volume (mm Tumor Tumor volume (mm Tumor 0 0 * 10 15 20 25 20 30 40 50 60 70 C Time post-implantation (days) Time post-implantation (days) PBS huOA-5D5 D Met Ki67 300 100 250 80

200 60 100 µM 100 µM 150 40 LY2875358 MM-131 100 Met H-score 50 20 0 0 % Ki67-positive tumor cells

Vehicle Vehicle 100 µM 100 µM MM-131 MM-131 Y2875358 huOA-5D5LY2875358 huOA-5D5L

Fig. 5. MM-131 inhibits tumor growth in vivo. NCI-H358 cells expressing human HGF (NCI-H358–HGF) (A)orMET-amplified cells (SNU-5) (B) were implanted s.c. in athymic nude (nu/nu) mice. Once tumor volumes reached 100–200 mm3, mice were randomized and dosed weekly with MM-131 (12 mg/kg), huOA-5D5 (10 mg/kg), LY2875358 (15 mg/kg), or vehicle. Tumor volume data plotted are mean (n = 8inA and n = 7inB) and SEM. (A) Final tumor volumes in MM-131– treated animals were significantly smaller than in vehicle-treated (**P = 0.0014), huOA-5D5–treated (***P = 0.0003), and LY2875358-treated (*P = 0.0321) groups. (B) In SNU-5 xenografts, final tumor volumes in MM-131–treated animals were significantly smaller than in vehicle-treated (****P < 0.0001), huOA-5D5– treated (***P = 0.0001), and LY2875358-treated (*P = 0.0331) treated groups. A complete response was observed in 100% (seven of seven) of MM-131– treated animals, 71% (five of seven) of LY2875358-treated animals, and 0% (zero of seven) of huOA-5D5– or vehicle-treated animals. Tumor volume data plotted are mean (n = 7) and SEM. (C) MM-131 inhibits Met activity in vivo. Representative images of pMet staining in SNU-5 tumors harvested 24 h after

treatment with the indicated antibodies. (D) Quantification of total Met and Ki67 in SNU-5 tumors harvested 24 h after treatment with the indicated an- PHARMACOLOGY tibodies. Scatter plots show all tumors (n = 4 per treatment) and SD.

via transcriptional up-regulation during tumor progression by (OS) in unselected patients, significant improvements were ob- microenvironmental factors such as hypoxia, therapeutic doses of ra- served in Met-positive patients, defined as having moderate (2+) diation, and exposure to HGF, which itself can activate Met to strong (3+) staining by IHC in ≥50% of tumor cells (59). transcription (9, 53, 54). HGF is frequently expressed in the acti- Unfortunately, this result was not repeated in a similarly designed vated stroma of primary tumors, and its expression has been linked phase 3 study (METLung), where the trial was halted early for to increased invasiveness and dissemination of tumor cells (4). In futility (60). More recently, onartuzumab was studied in a phase addition, elevated levels of HGF and/or Met can confer resistance 3 trial in metastatic HER2-negative, Met-positive gastroesopha- to traditional therapies. Given the role of Met in oncogene geal cancer in combination with mFOLFOX6 (METGastric), but “addiction,” as a prometastatic factor, and as a mediator of resistance, enrollment was stopped early due to negative final results in a Met is an attractive therapeutic target in many cancers. Importantly, similar phase 2 study (61). As with the METLung study, patients these roles are not mutually exclusive and can evolve during tumor in the METGastric study were selected based on IHC staining for progression as tumor cells interact with their microenvironment. Met, although, in this instance, the threshold for Met positivity Optimal targeting of Met requires activity in both the HGF- was set lower (IHC: 1+,2+,or3+). independent (MET-amplified) and HGF-dependent settings. Eli Lilly’s emibetuzumab (LY2875358), a bivalent anti-Met Early clinical development of Met TKIs, most notably tivantinib, antibody, was also recently studied in a phase 2 trial, either as had disappointing results (55, 56), although questions were later monotherapy or in combination with erlotinib in patients with raised concerning the target selectivity of tivantinib (15). Although many of the early Met TKI trials were performed in unselected patient populations, secondary analyses in patients with high levels Gastric of Met [as determined by immunohistochemistry (IHC)] showed CRC NSCLC Met Met Met signs of increased benefit. Subsequent to the phase 3 failure of HGFF HGF HGFF tivantinib in NSCLC (57), more recent development of Met TKIs 33% 33% 16% 30% has largely focused on patients with MET-amplified tumors. Al- 35% 42% though promising response rates have been observed in this setting, resistance typically emerges relatively quickly and can often be attributed to tumor microenvironment-derived HGF (21) or mutations in the tyrosine kinase domain of Met (58). EpCAM EpCAM EpCAM Antibody-based therapeutics offer an attractive alternative to TKIs, but the clinical development of anti-Met antibodies has Fig. 6. Met, EpCAM, and HGF are coexpressed in tumors. Primary tumor had limited success. Over the past decade, several anti-Met an- samples were collected from patients with CRC, gastric cancer, and NSCLC. tibody therapeutics have been studied in NSCLC (mostly in Tumors were stained for Met and EpCAM by IHC and for HGF by RNA in situ combination with EGFR TKIs) and gastric cancer, where Met is hybridization. Venn diagrams represent the percentage of tumors staining frequently activated by overexpression or gene amplification. positive for Met (light blue), EpCAM (red), and HGF (yellow). Percent coexpression of Met, EpCAM, and HGF is shown in dark blue (35%, 33%, and Genentech/Roche’s MetMAb/huOA-5D5 seemed promising in a ± 16% in CRC, gastric cancer, and NSCLC, respectively); percent coexpression of randomized phase II trial of erlotinib onartuzumab in patients Met and EpCAM (in the absence of HGF) is shown in purple (42%, 33%, and with chemorefractory NSCLC (59). Although the addition of 30% in CRC, gastric cancer, and NSCLC, respectively). Data used to generate onartuzumab to erlotinib did not increase PFS or overall survival Venn diagrams can be found in SI Appendix, Table S6.

Casaletto et al. PNAS | April 9, 2019 | vol. 116 | no. 15 | 7539 Downloaded by guest on October 2, 2021 Met-positive, metastatic NSCLC who had acquired resistance to absence of findings suggestive of liver cell damage, as well as erlotinib. In this case, Met positivity was defined as having the role of HGF/Met in maintaining liver homeostasis (67), moderate (2+) to strong (3+) staining in ≥10% of tumor cells. these toxicities likely arise from Met inhibition in hepato- Unfortunately, this trial also did not meet its primary end points, cytes. As hepatocytes are EpCAM-negative, the design of which were based on objective response rates (ORRs) (62). MM-131 may provide a wider therapeutic window in which Despite ongoing efforts, anti-Met antibody therapeutics remain to optimize dosing. Given the increase in potency afforded unsuccessful in late-stage clinical development. Here, we submit by EpCAM targeting, it is reasonable to expect that MM- 131 could be dosed at a level that permits inhibition of Met four reasons that may underlie these outcomes: in EpCAM-positive tumors cells while sparing EpCAM- i) Insufficient potency. Onartuzumab was designed as a mono- negative hepatocytes. Notably, there were no gross patho- valent antibody to prevent agonism. This reduction in logical findings or changes in clinical pathology parameters valency results in decreased potency, leading to insufficient in cynomolgus monkeys treated with up to 60 mg/kg of MM- inhibition of HGF-dependent Met activation. In addition, 131. In addition, Finisguerra et al. (68) recently reported monovalency removes the ability to down-regulate Met, ren- that inhibition of Met on immune cells may negatively im- dering onartuzumab unable to inhibit ligand-independent pact their antitumor properties, and that this may be an (MET-amplified) tumors (Figs. 3C,4B, and 5B). To circum- underlying cause of failure for some anti-Met therapies. vent this limitation, we designed MM-131 as a bispecific Data from the single-cell RNA-sequencing study published antibody: monovalent for Met, but also able to bind the by Tirosh et al. (69) indicate that they have little to no highly expressed tumor antigen EpCAM. This strategy re- EpCAM expression, suggesting that treatment with MM- stores potency in the ligand-dependent setting (Figs. 1H and 131 will not affect the antitumor immune response. 2B) and restores the ability to down-regulate Met in the ligand-independent (MET-amplified) setting (SI Appendix, In summary, we have shown that MM-131 is a purely antagonistic Fig. S6A). Notably, patients in both the METLung and anti-Met antibody that potently inhibits Met signaling in both ligand- METGastric studies were selected based on Met positivity, dependent and ligand-independent EpCAM-positive tumor cells by but those tumors that were ligand-independent, a setting in blocking HGF binding and inducing receptor down-regulation. which onartuzumab has no preclinical activity (Figs. 4B and These mechanisms of action lead to inhibition of cancer cell pro- 5B), were not excluded (60, 61). liferation and migration in vitro and inhibition of tumor growth in ii) Tumor-derived HGF was not considered. Although plasma vivo. MM-131 has a favorable pharmacokinetic and toxicity profile HGF has been measured in several studies, HGF produced and was designed to limit on-target toxicities reported for other Met inhibitors. These preclinical data support the evaluation of MM- in an autocrine manner or, more commonly, by activated MET stroma has not been considered. This likely stems from a 131 in patients with -amplified, Met-overexpressed, or Met- historical lack of qualified assays to measure tumor-derived positive/HGF-positive tumors that also express EpCAM. HGF. It is unknown whether local production of HGF is Methods critical for Met activity in the ligand-dependent setting, Cell Lines. Cell lines were obtained from the American Type Culture Collection, but it deserves attention, given the prevalence of HGF in National Cancer Institute Development Therapeutics Program, German Collec- human tumors and the predictive value it may hold (4, 11). tion of Microorganisms and Cell Cultures (Leibniz Institute, Braunschweig, Notably, onartuzumab is only active in the ligand-dependent Germany), or Sigma–Aldrich. HCC827-GR5 cells were kindly provided by setting, yet tumoral ligand levels were not reported in the P. Jänne, Dana–Farber Cancer Institute, Boston, MA. Details can be found in phase 2 or phase 3 trials (59–61). Similarly, bivalent anti- SI Appendix, Supplemental Methods. All cell lines were maintained bodies such as emibetuzumab (LY2875358) may agonize according to the source cell bank’s recommendations or in RPMI supple- Met signaling in the absence of ligand, as exemplified by mented with 10% FBS and 1% penicillin/streptomycin at 37 °C and 5% CO2. its promotion of migration in the absence of HGF (Fig. To generate EpCAM knockdown cell lines, cells were transduced with Mission – 4D and SI Appendix, Fig. S7D), yet tumoral HGF levels were shRNA lentiviral particles against EpCAM (SHCLNV-NM_002354; Sigma Aldrich). Mission nontarget shRNA control lentiviral particles (SHC016V) were also not reported in the emibetuzumab trials (62, 63). iii used to generate control cell lines. To generate HGF-expressing cell lines, cells ) Clinical studies typically focus on inhibition of tumor growth were transduced with lentiviral particles expressing human HGF (LPP-A0820- but rarely evaluate another major role of HGF/Met signal- Lv105-100-S; GeneCopoeia) or control viral particles (LPP-NEG-Lv105-025; ing, that of promoting tumor cell invasion and dissemina- GeneCopoeia). Lentiviral transductions were carried out in the presence of 4 tion. To date, early-stage clinical trials evaluating Met μg/mL polybrene, and cells were selected in 1–2 μg/mL puromycin for several inhibitors have used the well-established surrogate end weeks before use in experiments. points of PFS and ORR to predict OS. The inclusion of metastasis-free survival, a measure of how long a patient Expression and Purification of Antibodies. During antibody engineering, an- remains on therapy before developing new lesions, may be tibodies were either stably transfected into CHO-K1 cells or transiently a more appropriate way to evaluate a Met inhibitor in the transfected into 293F cells and harvested as described (70). Supernatants were incubated with MabSelect (GE Healthcare) while shaking at 20 °C for early stages of clinical development. Additionally, trials 1 h. Supernatant and resin were loaded onto a column (BioRad), and a designed specifically to evaluate prevention of metastasis vacuum was applied to separate supernatant from resin. The column was in patients with premetastatic disease may also be better washed with a high-conductivity buffer (PBS + 0.5 M NaCl). Antibodies were matched to Met biology than trials focused on patients with eluted with 0.1 M acetic acid, neutralized with 1 M Tris buffer (pH 8.0), filter- late-stage, metastatic disease. Notably, MM-131 inhibited sterilized, and dialyzed against PBS overnight at 4 °C. HGF-mediated tumor cell migration more effectively than MM-131 was stably expressed in a CHO-K1-GS cell line. Briefly, cells were either huOA-5D5 or LY2875358 (Fig. 4C and SI Appendix, electroporated with linearized DNA, transferred to a shaker flask, and in- Fig. S7C) and did not promote migration in the absence of cubated at 37 °C (120 rpm, 5% CO2). The next day, cells were centrifuged and HGF, as was observed with both huOA-5D5 and LY2875358 suspended in glutamine-free selection media containing L-methionine sul- (Fig. 4D and SI Appendix, Fig. S7D). It will be interesting to phoximine, distributed in 96-well plates at varying concentrations, and incubated at 37 °C. After 3–4 wk, the resulting minipools were screened for see if this preclinical effect on cell migration translates into expression, and the top three pools were combined and adapted into Lonza an antimetastatic effect in patients. proprietary media. Adapted cells were seeded into media at a density of 2 × iv 5 ) On-target toxicity. Two of the common toxicities reported in 10 cells per milliliter, and feeds were added per manufacturer recommen- clinical trials of antibody-based Met pathway inhibitors are dations. On day 14, cells were harvested by centrifugation and the super- peripheral edema and hyperalbuminemia (64–66). Given the natant was filtered using a 0.22-μm filter.

7540 | www.pnas.org/cgi/doi/10.1073/pnas.1819085116 Casaletto et al. Downloaded by guest on October 2, 2021 The sequences for huOA-5D5 (29) and LY2875358 (US Patent 9201074; ref. 5,000 cells per well in growth medium supplemented with 4% FBS. Following 71) were stably expressed in CHO-K1 cells as described (70). ABT-700 (US Patent 48 h of incubation to allow for spheroid formation, cells were treated with 8545839; ref. 72), MM-131EpCAMmut,andMM-131Metmut were transiently 1 nM HGF and/or antibodies, as indicated. Treatment medium was replen- expressed using the Expi293 expression system (Life Technologies). Mutations ished after 48 h. Following 96 h of treatment, cell viability was determined by were made in the paratope to eliminate binding to EpCAM and Met. On day 6, incubation with CellTiter-Glo (Promega) for 10 min, followed by luminescent cells were harvested by centrifugation and the supernatant was filtered using readout on an EnVision plate reader (PerkinElmer). Values were normalized a0.22-μM filter. All proteins were purified using protein A chromatography to vehicle-treated cells in the presence or absence of 1 nM HGF (as indicated), and characterized with SEC as described (70). All antibodies were confirmed to and dose–response curves were determined by a four-parameter fit using be greater than 90% monomeric species. Final concentrations were de- GraphPad Prism software. termined using A280 values and theoretical extinction coefficients. Cell Migration Assays. Cells were grown to confluence on 96-well ImageLock Affinity Measurements. All reagents were equilibrated, and samples were plates (Essen Bioscience) and wounded with a 96-pin wound-making tool brought to room temperature before the biolayer interferometry assay (WoundMaker; Essen Bioscience). Detached cells were removed by a serum- (ForteBio). Streptavidin biosensors were hydrated for 10 min in protein-free free medium wash. Serum-free medium containing 1 nM HGF and/or anti- blocking buffer (Pierce), followed by 10 min in PBS. Assays were run using bodies was added to wells and wounds were imaged for 24 h with the the Octet software and procedure per the manufacturer’s instructions. The IncuCyte ZOOM live cell imager (Essen Bioscience). The relative wound density assay consisted of the following steps: 10-min equilibration, 2-min biotinylated (density of cells inside relative to outside the wound) was quantified at 2-h μ antibody loading (10 g/mL), 1-min baseline stabilization, and 3-min in- intervals, and migration rates were calculated from these data. Values were – cubation with the antigen listed (200 12.5 nM), followed by 5-min measure- normalized to those of cells in the absence of antibodies and treated with or ment of the dissociation. PBS was used as the matrix throughout. Data were without 1 nM HGF. Dose–response curves were determined by a four- analyzed and processed with Octet data analysis software to determine kinetic parameter fit using GraphPad Prism software. parameters [Kd, on-rate constant (kon), and off-rate constant (koff)]. Tumor Xenografts. NCI-H358–HGF, HCC827-HGF, SNU-5, MKN-45, and NCI- Met and EpCAM Cell Surface Quantification by Quantitative Fluorescence- H441 cells were implanted s.c. into the flank of female athymic nude Activated Cell Sorting. Cells were seeded at least 48 h before harvest. Cells (nu/nu) mice. NCI-H1993 cells were implanted into female BALB/c nude mice. For were detached with Accutase (A1110501; ThermoFisher Scientific), washed, MKN-45 xenografts, 2 × 106 cells were implanted; 5 × 106 cells were implanted and stained with fluorescently labeled anti-Met (ABT-700) or anti-EpCAM for all other models. Tumor volumes [V = (π/6) × L × W2] were measured with (EBA-1; BD Biosciences) antibodies. Quantum Simply Cellular anti-human calipers. When volumes reached 100–200 mm3, animals were randomized IgG beads (Bangs Laboratories) were similarly labeled. Labeled beads and according to tumor volume and body weight. Animals were dosed weekly cells were analyzed by flow cytometry. Four populations of beads with with MM-131 (12 mg/kg), huOA-5D5 (10 mg/kg), or LY2875358 (15 mg/kg) (all

varying antibody-binding capacities were used as standards to quantify the PHARMACOLOGY equimolar equivalents), or with PBS via i.p. administration. All animal studies number of surface receptors per cell. were conducted in accordance with the Guide for the Care and Use of Lab- oratory Animals (76). The Institutional Animal Care and Use Committee at ELISAs. Cell-based ELISAs were performed using standard techniques. De- Merrimack Pharmaceuticals reviewed and approved all animal protocols. tailed protocols are provided in SI Appendix, Supplemental Methods. IHC and RNA in-Situ Hybridization. Methods for IHC and RNA in situ hybrid- HGF-Binding Assay. Biolayer interferometry (ForteBio) was used to measure ization are described in SI Appendix, Supplemental Methods. the HGF-blocking properties of anti-Met antibodies. Antibodies were loaded onto anti-human IgG Fc capture biosensors and incubated with 200 nM Mechanistic Computational Model. A mechanistic computational model was human Met. After 2 min, 200 nM human HGF was added, followed by ad- constructed to assess differential activities of the Met- and EpCAM-binding dition of PBS after another 2 min. capabilities of MM-131 and Met binding of huOA-5D5 on pMet and pAkt signaling and their dependency on Met and EpCAM expression of the cell MM-131 Structure Determination. Methods for crystallization, data collection, lines. The topology of the model includes antibody–receptor interactions, and structure determination of MM-131 are described in SI Appendix, ligand-induced formation of phosphorylated dimers, and signaling to key Supplemental Methods (73–75). downstream proteins. Values of the kinetic parameters in the model were constrained by fitting the model simulation to experimental dose–response Degradation/Signaling Assays and Immunoblotting. For degradation assays, and time-course data, for a technical description of the approach (77). The cells were starved overnight in media containing 0.5% FBS, incubated with reactions in the model are represented as a system of ordinary differential 100 nM indicated antibodies or PBS for 24 h, lysed in radioimmunoprecipitation equations. Model construction and simulation were performed with assay buffer (89901; ThermoFisher Scientific) supplemented with protease and MATLAB (MathWorks) and the D2D modeling framework (http://www. phosphatase inhibitors as above, and analyzed by immunoblotting. For sig- data2dynamics.org/). The model topology is described in SI Appendix, naling assays, cells were starved overnight in media containing 0.5% FBS, Tables S7 and S8. All experimental data and code used for model con- incubated with 1 nM HGF and/or MM-131 as indicated, and lysed and analyzed struction and parameter calibration are available at http://www. as above. Antibodies to pMet (Y1234/35), Met, pAkt (S473), Akt, pGab1 (Y659), data2dynamics.org/. Gab1, pErk1/2, Erk1/2, EpCAM, actin, and histone H3 were from Cell Signaling Technology. Bands were quantitated using a LI-COR Odyssey imaging system. ACKNOWLEDGMENTS. We thank Victoria Rimkunas for assistance with IHC staining and scoring, Johanna Lahdenranta for assistance with HALO Cell Proliferation Assays. To perform cell proliferation assays in 3D spheroid software analysis, Ken Olivier for design and analysis of the toxicology and cultures, cells were plated into 96-well, low-binding, multispheroid culture toxicokinetic studies, and Jae Kim for assistance with pharmacokinetic plates (NCP-LH96-10; Scivax Life Sciences) at densities between 3,000 and analysis and modeling.

1. Trusolino L, Bertotti A, Comoglio PM (2010) MET signalling: Principles and functions in 7. Organ SL, Tsao M-S (2011) An overview of the c-MET signaling pathway. Ther Adv development, organ regeneration and cancer. Nat Rev Mol Cell Biol 11:834–848. Med Oncol 3(Suppl 1):S7–S19. 2. Gherardi E, Birchmeier W, Birchmeier C, Vande Woude G (2012) Targeting MET in 8. Raghav KPS, Gonzalez-Angulo AM, Blumenschein GR, Jr (2012) Role of HGF/MET axis in cancer: Rationale and progress. Nat Rev Cancer 12:89–103. resistance of lung cancer to contemporary management. Transl Lung Cancer Res 1:179–193. 3. Ferracini R, et al. (1995) The Met/HGF receptor is over-expressed in human osteosar- 9. De Bacco F, et al. (2011) Induction of MET by ionizing radiation and its role in ra- comas and is activated by either a paracrine or an autocrine circuit. Oncogene 10: dioresistance and invasive growth of cancer. J Natl Cancer Inst 103:645–661. 739–749. 10. Engelman JA, et al. (2007) MET amplification leads to gefitinib resistance in lung 4. Matsumoto K, Nakamura T (2006) Hepatocyte growth factor and the Met system as a cancer by activating ERBB3 signaling. Science 316:1039–1043. mediator of tumor-stromal interactions. Int J Cancer 119:477–483. 11. Siegfried JM, et al. (1998) The clinical significance of hepatocyte growth factor for 5. Grugan KD, et al. (2010) Fibroblast-secreted hepatocyte growth factor plays a func- non-small cell lung cancer. Ann Thorac Surg 66:1915–1918. tional role in esophageal squamous cell carcinoma invasion. Proc Natl Acad Sci USA 12. Nakajima M, et al. (1999) The prognostic significance of amplification and over- 107:11026–11031. expression of c-met and c-erb B-2 in human gastric carcinomas. Cancer 85:1894–1902. 6. Comoglio PM, Giordano S, Trusolino L (2008) Drug development of MET inhibitors: 13. Zeng Z-S, et al. (2008) c-Met gene amplification is associated with advanced stage Targeting oncogene addiction and expedience. Nat Rev Drug Discov 7:504–516. colorectal cancer and liver metastases. Cancer Lett 265:258–269.

Casaletto et al. PNAS | April 9, 2019 | vol. 116 | no. 15 | 7541 Downloaded by guest on October 2, 2021 14. Go H, et al. (2010) High MET gene copy number leads to shorter survival in patients 47. Birchmeier C, Birchmeier W, Gherardi E, Vande Woude GF (2003) Met, metastasis, with non-small cell lung cancer. J Thorac Oncol 5:305–313. motility and more. Nat Rev Mol Cell Biol 4:915–925. 15. Katayama R, et al. (2013) Cytotoxic activity of tivantinib (ARQ 197) is not due solely to 48. Rong S, et al. (1992) Tumorigenicity of the met proto-oncogene and the gene for c-MET inhibition. Cancer Res 73:3087–3096. hepatocyte growth factor. Mol Cell Biol 12:5152–5158. 16. Gan HK, et al. (2015) cMet: Results in papillary renal cell carcinoma of a phase I study 49. Miller CT, et al. (2006) Genomic amplification of MET with boundaries within fragile of AZD6094/volitinib leading to a phase 2 clinical trial with AZD6094/volitinib in pa- site FRA7G and upregulation of MET pathways in esophageal adenocarcinoma. tients with advanced papillary renal cell cancer (PRCC). J Clin Oncol 33(Suppl 7):487 Oncogene 25:409–418. (abstr). 50. Bean J, et al. (2007) MET amplification occurs with or without T790M mutations in 17. Soo R, et al. (2015) Highly selective c-Met inhibitor tepotinib plus gefitinib is active in EGFR mutant lung tumors with acquired resistance to gefitinib or erlotinib. Proc Natl Asian patients with c-Met+ NSCLC. Ann Oncol 26:125–147. Acad Sci USA 104:20932–20937. 18. Bauer TM, et al. (2016) Phase (Ph) I study of the safety and efficacy of the cMET in- 51. Bardelli A, et al. (2013) Amplification of the MET receptor drives resistance to anti- hibitor capmatinib (INC280) in patients (pts) with advanced cMET+ non-small cell lung EGFR therapies in colorectal cancer. Cancer Discov 3:658–673. cancer (NSCLC). J Thorac Oncol 11(Suppl):S257–S258. 52. Turke AB, et al. (2010) Preexistence and clonal selection of MET amplification in EGFR 19. Lennerz JK, et al. (2011) MET amplification identifies a small and aggressive subgroup mutant NSCLC. Cancer Cell 17:77–88. of esophagogastric adenocarcinoma with evidence of responsiveness to crizotinib. 53. Pennacchietti S, et al. (2003) Hypoxia promotes invasive growth by transcriptional J Clin Oncol 29:4803–4810. activation of the met protooncogene. Cancer Cell 3:347–361. 20. Vassal G, et al. (2015) Activity of crizotinib in relapsed MET amplified malignancies: 54. Boccaccio C, Gaudino G, Gambarotta G, Galimi F, Comoglio PM (1994) Hepatocyte Results of the French AcSé Program. J Clin Oncol 33(Suppl 15):2595 (abstr). growth factor (HGF) receptor expression is inducible and is part of the delayed-early – 21. Pennacchietti S, et al. (2014) Microenvironment-derived HGF overcomes genetically response to HGF. J Biol Chem 269:12846 12851. determined sensitivity to anti-MET drugs. Cancer Res 74:6598–6609. 55. Sequist LV, et al. (2011) Randomized phase II study of erlotinib plus tivantinib versus 22. Prat M, Crepaldi T, Pennacchietti S, Bussolino F, Comoglio PM (1998) Agonistic erlotinib plus placebo in previously treated non-small-cell lung cancer. J Clin Oncol 29: – monoclonal antibodies against the Met receptor dissect the biological responses to 3307 3315. HGF. J Cell Sci 111:237–247. 56. Kang Y-K, et al. (2014) A phase II trial of a selective c-Met inhibitor tivantinib (ARQ 23. Pacchiana G, et al. (2010) Monovalency unleashes the full therapeutic potential of the 197) monotherapy as a second- or third-line therapy in the patients with metastatic – DN-30 anti-Met antibody. J Biol Chem 285:36149–36157. gastric cancer. Invest New Drugs 32:355 361. 24. Moores SL, et al. (2016) A novel bispecific antibody targeting EGFR and cMet is ef- 57. Scagliotti G, et al. (2015) Phase III multinational, randomized, double-blind, placebo- fective against EGFR inhibitor-resistant lung tumors. Cancer Res 76:3942–3953. controlled study of tivantinib (ARQ 197) plus erlotinib versus erlotinib alone in previously 25. Choi H-J, Kim Y-J, Lee S, Kim Y-S (2013) A heterodimeric Fc-based bispecific antibody treated patients with locally advanced or metastatic nonsquamous non-small-cell lung – simultaneously targeting VEGFR-2 and Met exhibits potent antitumor activity. Mol cancer. JClinOncol33:2667 2674. Cancer Ther 12:2748–2759. 58. Bahcall M, et al. (2016) Acquired METD1228V mutation and resistance to MET in- – 26. Herlyn M, Steplewski Z, Herlyn D, Koprowski H (1979) Colorectal carcinoma-specific hibition in lung cancer. Cancer Discov 6:1334 1341. 59. Spigel DR, et al. (2013) Randomized phase II trial of onartuzumab in combination antigen: Detection by means of monoclonal antibodies. Proc Natl Acad Sci USA 76: with erlotinib in patients with advanced non-small-cell lung cancer. J Clin Oncol 31: 1438–1442. 4105–4114. 27. van der Gun BTF, et al. (2010) EpCAM in carcinogenesis: The good, the bad or the 60. Spigel DR, et al. (2014) Onartuzumab plus erlotinib versus erlotinib in previously ugly. Carcinogenesis 31:1913–1921. treated stage IIIb or IV NSCLC: Results from the pivotal phase III randomized, multi- 28. Schnell U, Cirulli V, Giepmans BNG (2013) EpCAM: Structure and function in health center, placebo-controlled METLung (OAM4971g) global trial. J Clin Oncol 32(Suppl and disease. Biochim Biophys Acta 1828:1989–2001. 5):8000 (abstr). 29. Merchant M, et al. (2013) Monovalent antibody design and mechanism of action of 61. Shah MA, et al. (2015) METGastric: A phase III study of onartuzumab plus mFOLFOX6 onartuzumab, a MET antagonist with anti-tumor activity as a therapeutic agent. Proc in patients with metastatic HER2-negative (HER2−) and MET-positive (MET+) ade- Natl Acad Sci USA 110:E2987–E2996. nocarcinoma of the stomach or gastroesophageal junction (GEC). J Clin Oncol 33 30. Harms BD, Kearns JD, Iadevaia S, Lugovskoy AA (2014) Understanding the role of (Suppl):4012 (abstr). cross-arm binding efficiency in the activity of monoclonal and multispecific thera- 62. Camidge DR, et al. (2016) A randomized, open-label, phase 2 study of emibetuzumab peutic antibodies. Methods 65:95–104. plus erlotinib (LY+E) and emibetuzumab monotherapy (LY) in patients with acquired 31. Ridgway JB, Presta LG, Carter P (1996) ‘Knobs-into-holes’ engineering of antibody resistance to erlotinib and MET diagnostic positive (MET Dx+) metastatic NSCLC. J Clin CH3 domains for heavy chain heterodimerization. Protein Eng 9:617–621. Oncol 34(Suppl):9070 (abstr). 32. Oganesyan V, Gao C, Shirinian L, Wu H, Dall’Acqua WF (2008) Structural character- 63. Rosen LS, et al. (2017) A first-in-human phase I study of a bivalent MET antibody, ization of a human Fc fragment engineered for lack of effector functions. Acta emibetuzumab (LY2875358), as monotherapy and in combination with erlotinib in Crystallogr D Biol Crystallogr 64:700–704. advanced cancer. Clin Cancer Res 23:1910–1919. 33. Zhang Y, Skolnick J (2005) TM-align: A protein structure alignment algorithm based 64. Morley R, et al. (2015) Safety of onartuzumab in patients with solid tumors: Experi- on the TM-score. Nucleic Acids Res 33:2302–2309. ence to date from the onartuzumab clinical trial program. PLoS One 10:e0139679. 34. Davies AM, Jefferis R, Sutton BJ (2014) Crystal structure of deglycosylated human 65. Strickler JH, et al. (2014) Phase 1, open-label, dose-escalation, and expansion study of IgG4-Fc. Mol Immunol 62:46–53. ABT-700, an anti-C-met antibody, in patients (pts) with advanced solid tumors. J Clin 35. Davies AM, et al. (2014) Structural determinants of unique properties of human IgG4- Oncol 32(Suppl 5):2507 (abstr). – Fc. J Mol Biol 426:630 644. 66. Cunningham D, et al. (2015) Phase III, randomized, double-blind, multicenter, placebo 36. Elliott JM, et al. (2014) Antiparallel conformation of knob and hole aglycosylated (P)-controlled trial of rilotumumab (R) plus epirubicin, cisplatin and capecitabine half-antibody homodimers is mediated by a CH2-CH3 hydrophobic interaction. J Mol (ECX) as first-line therapy in patients with advanced MET-positive (pos) gastric or – Biol 426:1947 1957. gastroesophageal junction (G/GEJ): RILOMET-1 study. J Clin Oncol 33(Suppl):4000 37. Wyrzucki A, et al. (2014) Alternative recognition of the conserved stem epitope in (abstr). influenza A virus hemagglutinin by a VH3-30-encoded heterosubtypic antibody. 67. Huh C-G, et al. (2004) Hepatocyte growth factor/c-met signaling pathway is required – J Virol 88:7083 7092. for efficient liver regeneration and repair. Proc Natl Acad Sci USA 101:4477–4482. 38. Landgraf KE, et al. (2014) An allosteric switch for pro-HGF/Met signaling using zy- 68. Finisguerra V, et al. (2015) MET is required for the recruitment of anti-tumoural – mogen activator peptides. Nat Chem Biol 10:567 573. neutrophils. Nature 522:349–353. 39. Krissinel E, Henrick K (2007) Inference of macromolecular assemblies from crystalline 69. Tirosh I, et al. (2016) Dissecting the multicellular ecosystem of metastatic melanoma – state. J Mol Biol 372:774 797. by single-cell RNA-seq. Science 352:189–196. 40. Liu L, et al. (2014) LY2875358, a neutralizing and internalizing anti-MET bivalent 70. Xu L, et al. (2013) Rapid optimization and prototyping for therapeutic antibody-like antibody, inhibits HGF-dependent and HGF-independent MET activation and tumor molecules. MAbs 5:237–254. growth. Clin Cancer Res 20:6059–6070. 71. Davies J, Liu L, Lu J; Eli Lilly and Co. (2015) Anti-C-Met Antibodies. US Patent US 41. Stamos J, Lazarus RA, Yao X, Kirchhofer D, Wiesmann C (2004) Crystal structure of the 9,201,074 B2 (December 1, 2015). HGF beta-chain in complex with the Sema domain of the Met receptor. EMBO J 23: 72. Goetsch L, Wurch T, Bes C; Pierre Fabre Medicament (2012) Anti-C-Met Antibody. US 2325–2335. Patent US 8,545,839 B2 (October 1, 2013). 42. Roovers RC, et al. (1998) High-affinity recombinant phage antibodies to the pan- 73. Casaletto JB, et al. (2019) Heteromeric tandem IgG4/IgG1 Fc. Protein Data Bank. carcinoma marker epithelial glycoprotein-2 for tumour targeting. Br J Cancer 78: Available at http://www.rcsb.org/pdb/search/structidSearch.do?structureId=6HYG. 1407–1416. Deposited October 21, 2018. 43. Willuda J, et al. (1999) High thermal stability is essential for tumor targeting of an- 74. Casaletto JB, et al. (2019) Crystal structure of Sema domain of the Met receptor in tibody fragments: Engineering of a humanized anti-epithelial glycoprotein-2 (epi- complex with FAB. Protein Data Bank. Available at http://www.rcsb.org/pdb/search/ thelial cell adhesion molecule) single-chain Fv fragment. Cancer Res 59:5758–5767. structidSearch.do?structureId=6I04. Deposited October 25, 2018. 44. Pavsicˇ M, Guncarˇ G, Djinovic-Carugo K, Lenarciˇ cˇ B (2014) Crystal structure and its 75. Casaletto JB, et al. (2019) Crystal structure of EpCAM in complex with scFv. Protein bearing towards an understanding of key biological functions of EpCAM. Nat Data Bank. Available at http://www.rcsb.org/pdb/search/structidSearch.do? Commun 5:4764. structureId=6I07. Deposited October 25, 2018. 45. Münz M, et al. (2010) Side-by-side analysis of five clinically tested anti-EpCAM 76. National Research Council (2011) Guide for the Care and Use of Laboratory Animals monoclonal antibodies. Cancer Cell Int 10:44. (National Academies Press, Washington, DC), 8th Ed. 46. Wang J, et al. (2016) Anti-c-Met monoclonal antibody ABT-700 breaks oncogene 77. Raue A, et al. (2015) Data2Dynamics: A modeling environment tailored to parameter addiction in tumors with MET amplification. BMC Cancer 16:105. estimation in dynamical systems. Bioinformatics 31:3558–3560.

7542 | www.pnas.org/cgi/doi/10.1073/pnas.1819085116 Casaletto et al. Downloaded by guest on October 2, 2021